What If The Expected Actually Happens?

Part 1: “Dewey Defeats Truman”

In the early morning hours of July 15th, 1948, in a sweltering Philadelphia Convention Hall, Democratic delegates had just finished days of arguing over civil rights and whether they would — in fact — nominate the sitting U.S President Harry Truman as their candidate for President that year. Truman was slightly favored to be nominated and thus was scheduled to accept the nomination on July 14th at ten in the evening; but the rancorous convention, which included southern delegates walking out to back third party candidates, had been behind schedule. So, with barely anyone still tuning in, and a hall filled with tired, sweaty, and pessimistic delegates, Truman finally accepted his nomination. When he started his speech, he was a massive underdog to earn his own elected term to office.

After FDR’s death in 1945, Truman had ascended to the Presidency and guided the country through the final days of World War II. He quickly became massively popular with his approvals going into the high eighties in the aftermath of the war. But being in the shadow of FDR, economic troubles, same party in power fatigue, and other circumstances lead him to become a very unpopular President afterwards. By 1946, Republicans swept their way to a wave election that midterm year. Truman’s popularity went up a bit, but regressed to mean eventually, and going into the 1948 campaign he had approvals in the lower forties. Republicans for their part had tapped New York Governor Thomas Dewey, a man who four years prior had given FDR his toughest re-election battle ever – even while the venerated President enjoyed high “rally to the flag” support as he guided the country through war.

And now here Truman was, he had to fight for the nomination, his acceptance speech had been pushed past midnight, and the delegates looked like they were ready to finally end their winning streak of winning the White House. But Truman delivered a speech that to this day is ranked as perhaps the greatest acceptance speech ever given. He excoriated the Republican congress’ record, attacked them on what he knew his party had the edge on at the time, and then called for a special session of congress to call them on their bluff. Truman’s speech had the delegates cheering and standing by the end of it, ready to go through a wall for him, and the campaign rally of “Give ’em Hell, Harry!” was born.

Truman’s bluff worked in his favor, and he went on a cross-country train trip in what came to be known as his “Whistle Stop Tour.” He campaigned like his life depended on it, and went from being down thirteen in the Gallup August poll to being down by five with about two weeks left to go. Pollsters stopped polling the race then, expecting Dewey would win even as Truman closed the gap. For his part Dewey’s internal pollsters, a first for a campaign, had seen Truman’s comeback in their numbers — particularly in the then battleground west. Come election day, Truman won a competitive race, upsetting Dewey to get his own elected term, and grabbing a Chicago Tribune newspaper that called the race too soon for his challenger to mock his foes with – as one photographer caught what would be a defining image of American history. Truman’s infamous gloating photo is the image of the unexpected.

Truman’s upset, even seventy-two years later, remains a shadow over every election since then, with upsets being referred to as “Dewey defeats Truman moments.” It is the measuring stick for every electoral upset that has come since, even mentioned across the sea in other country’s elections when major upsets occur. It is akin to bringing up the Patriots’ comeback against the Falcons in Super Bowl 51 or the Detroit Redwings choking away a 3-0 Stanley Cup Finals lead to the Toronto Maple Leafs in 1942. And it came to mind when arguably the only Presidential election to see such a surprise since occurred in 2016 with Donald Trump’s victory.

So, why do I bring this all up? Well, as I’ve slowly but surely binge 1920s silent films to prepare for my next film piece, and as I await for Labor Day weekend to arrive before I write my final pre-election analysis of 2020, I see a trend that a lot of folks believe there’s a ton of “uncertainty” and potential surprises for this year’s election. From a potential 2016 repeat to a chaotic election night that turns into weeks or maybe even months. And honestly it is starting to become a bit of an annoyance for me, not because I think there’s anything wrong with keeping an open mind that perhaps we will see surprises, but because I’ve seen this pattern play out before in a lot of elections I’ve followed, both here in the U.S and over in the UK. So with every pundit trampling over each other to write the “We’ve been here before!” pieces or predictions of another 2000 debacle, I wanted to be daring (because I’ll have egg on my face if all these fears come true) and actually argue for the possibility that maybe, just maybe, the expected…actually happens and maybe even smoothly at that.

As you can see from this political cartoon of the time, Truman was a massive underdog in 1948, he still ended up winning

Part 2: The Arguments For The Unexpected

As I sit down and write this, the Republican Convention is set to happen with the President hoping it’ll give him the boost he needs as we head towards the homestretch of the campaign, with he still unpopular and his challenger leading on average by 8-10 points and hitting ~50%+ with a historically low amount of undecideds for this stage of the campaign. With Biden’s convention going off with very few hitches and having a well-received speech, many are wondering if this is the calm before the storm of a would-be Trump 2016-type comeback. After all, the President could hammer his challenger’s image effectively for four days of prime-time TV, tighten the race, and then end up with another close popular vote loss/close electoral college victory situation. Now I’ll expand on that in my next Election 2020 analysis, but we’re seeing a lot of forecasters and election analysts still filled with trepidation that we’re headed towards another 2016. Nate Silver released a forecast model where he purposely cranked up the levels of potential error, Dave Wasserman has been obsessively focused on the strengths and weaknesses of the message Biden has for the working class white voters, and others like Nate Cohn have argued with others on just how high Biden’s popular vote floor must be to win the electoral college.

And then of course there’s the mess with the mail ballots. On one hand, the COVID-19 pandemic has forced mail voting to skyrocket this year, causing concern that the ballot counting could lead to us taking days or a week to figure out a winner even if it ends up a moderate to big Biden win. Or perhaps maybe we end up in a month plus of legal drama because the race is close enough that every mail ballot count is so important. On the other hand, the President has politicized this situation so that a majority of mail voting will likely be done by Democrats and a majority of “day of” voting will be done by Republicans. Thus, even if Biden wins, early returns could look good enough for the President that he erroneously declares himself the winner too early, causing chaos when the race maybe goes the other way. After all, we saw a lot of conservative pundits on social media make this mistake in 2018 with the close Arizona Senate race and that was with typical ballot counting. And that’s before mentioning the mess with the USPS slow mail scandal the Trump administration and the USPS is dealing with at the moment.

So, we have arguments that we could see the presidential election change dramatically or have a major polling miss that leads to another Trump upset. We have concerns that mail ballot voting could lead to painfully slow counts that force us to wait too long to know who has won the main event. We also have a very real fear that the President could declare himself the winner too early, just to end up losing as mail ballots continue to be accounted for. Now that we know what all these points of chaos are, let me address them and make my (optimist) case for why perhaps things might not end up as chaotic as one might think…hopefully.

Another 2016 shocker is only one of many things folks seem to be worrying about

Part 3: What If Biden Wins As Expected?

2016’s polling miss has ABSOLUTELY affected the coverage of this race. I think that’s a good thing in some cases but has become a hindrance in a lot of others. If we had treated a candidate getting ~50%+ support with a 8-10 point lead like this in the past, a candidate who’s challenging a sitting President with an objectively bad 42% approval rating, with a recession and pandemic raging in the background, certain forecasters that today preach about “UnCeRtAiNtY” in the wake of 2016 would be tweeting about the media wanting a closer race than exists. If you lived through the daily 2012 punditry sphere on social media, Obama then had a weaker standing than Biden has now, and yet there was a very strong confidence among forecasters and analysts that the race was leaning his way. And in 2016, the confidence talk by a lot (not all to be fair) of these same folks lead to the disconnect between the actual polling data and the expectations of the race’s margins.

Now I don’t think there’s any great scandal in keeping your eyes out for a late surge for Trump or the possibility the RNC ends up a game changer (Throughout history conventions like the 1988 RNC, the 2004 RNC, and 2012 DNC have done wonders for parties in power who need a boost). However, given the incredibly low amount of undecideds, I don’t think its scandalous either to say the President has dug himself into quite a hole and that the window for a comeback is getting narrower and narrower.

Constantly sitting around and waiting for a comeback based on the last election, only for it to never come, has happened before. In 1988, then sitting Vice President George H.W Bush roared from a big deficit against then-Massachusetts Governor Michael Dukakis to be elected to the White House. Flash forward to 1992 and Republicans kept waiting and hoping for such a comeback to come against Clinton. It never did, past the race tightening a bit down the stretch but by then it was too little, too late. In 2012, Republicans seemed poised to flip the Senate even while their Presidential candidate struggled in his race, but after Labor Day weekend Democrats solidified themselves as favorites to keep the upper chamber. Two years later in 2014, Democrats were on my twitter feed then hoping to see a similar comeback, but it never came, and Republicans flipped the Senate as they had been favored to do so that midterm year.

Now, what about the worries of a major miss in polling? While understandable after 2016, I actually think this is a less likely outcome than an outright Trump comeback. For one, as I’ve explained a million times in previous articles and on social media, while the 2016 miss was a big hit for the perception of the industry, it was also statistically a small and typical miss that many have learned from. Pollsters are now weighting by education more, and more eyes are on the number of undecideds and what exact percentage of support the candidates are polling at. Pro and amateur election followers alike didn’t sit around and just accept failure, they learned from it and pollsters did as well. So, the likelihood of another miss in my mind is lower just off of that. Not to mention Biden’s leads (if they remain this way up until election day) should be big enough to survive a 2016-type miss, and again there’s less undecideds and his percentage of support is hitting ~50%+ in national and state battleground polling whereas Clinton had smaller leads and struggled to hit the mid-forties. Remember that while polling is never perfect, it followed up 2016 with correct calls in 2017, 2018, 2019, and in the presidential primaries from earlier in the year.

In 1952, four years after Truman’s upset, pollsters were so scared of their failure that even as Eisenhower lead consistently against Stevenson, they were questioning if they were right. In fact, Gallup, which had Ike with a commanding lead in their final poll, predicted a tighter race then their polling found in their last press release – Ike ended up winning easily and the polls beat the punditry. Over in the UK, there was a big upset when the Conservatives were able to hang on to a majority even though polls had Labour favored to form a minority to small majority government. This error in polling left the Tories in 1997, when they were down by landslide margins in the polls, to expect to overperform polls so much that perhaps the race might end up more competitive. When Blair’s landslide was announced and felt as the night went along, there were stories of Conservative “victory party” attendees watching the election returns with mouths agape, in tears, and even a shattered wine glass or two. Or even recently when 2018’s polling showed Democrats consistently favored by high single digits in the Generic Ballot, but the 2016 PTSD had some writing badly aged Summer articles on a coming GOP comeback. And of course, in the UK we can see what happened when in 2017 a 2015-type miss in the Tories’ favor was expected, and in 2019 when a 2017-type miss in Labour’s favor was expected – neither came to pass.

So yes, we can speculate all day that Joe Biden might end up choking away a big lead, we can perhaps worry that another big polling miss might occur, but I’ve seen this focus on the previous election before cloud expectations and cause what I like to call “needless shock”. There was no reason to be surprised when Bush couldn’t pull off the comeback again in 1992, when in 1952 Eisenhower lead all the way and won the election, when Labour had the landslide polls expected they’d have in 1997, or when Democrats bounced back from 2016 and were favored from start to finish to pull off what they did in 2018. You honestly have zero reason to be surprised if when the fog clears Joe Biden has won the election, and if he’s done it by the margins polls expected him to.

Even though Eisenhower lead Stevenson from wire to wire in 1952, 1948 was always in the back of many minds. Gallup went against their polling when they made their final prediction (Gallup used to make their own subjective final predictions)

Part 4: What If A Chaotic Election Night/Week Doesn’t Happen?

There is absolutely reason to be concerned about the smoothness of mail ballot counting, the USPS scandal, and I can tell you right now you will have to learn to be patient for some closer results because of the pandemic. I have no beef with any of these points made when someone worries about electoral returns chaos. However, I do think there’s a reason to be optimistic that things end up running much smoother than one might be worried about.

For one, the primaries gave some states a good chance to test out things and learn from them, this includes the issue of mail ballot rejections. I’m not saying there’s no gaps in the system, but most election administrators are trying hard and have the experience to make sure your vote is counted. In my state of Florida, the ballot counting has gone pretty smoothly as we’re known for our fast counts and experience with mail and absentee ballot counting. Second, if Biden maintains the lead he has now as I write this, he should be declared the winner on election night leaving us to figure out the margins he won by in the weeks afterwards. One major part of this is because of the already mentioned fast counting of Florida where Biden supposedly has bigger than expected margins in his lead there. If he wins the state pretty easily, we’ll know for sure earlier than one might expect, and that would mean the polls are probably about right. Again, this doesn’t mean there won’t be close calls that take time to figure out (I tend to think the Senate might take the longest to figure out), but there is a real possibility we know a lot of the most important stuff by the end of the first night.

I’ve seen these sorts of worries before. Going into election night 2004, after the debacle of 2000, there were fears of another popular vote/electoral college split that would cause months of legal battles — a 2000 on steroids sort of situation (Watch coverage of the night on YouTube, networks put a lot of man power to cover a lot of legal battles that didn’t end up mattering). Now 2004 was a very close race, but it was a landslide compared to those fears when we knew the next morning Bush had won, and the Republicans had gained seats in the House and Senate. Over in the UK, there was a lot of hunkering down expected for what was thought to be a potential for weeks of party coalition negotiations when polls closed for the 2015 race – the exit polls ended that in a flash. If come election night we actually know the winner and returns seem to be going smoothly, there’s no reason to be overly surprised. As for the potential for Trump to declare himself the winner too early? Unless the race is so easily won by Biden, he is declared the winner on election night, even I can’t give you optimism over that potential chaos.

The 2004 election had months of hype as another 2000 debacle in the making, while it was close we knew the winner quicker than some had anticipated

Part 5: More Likely Than Not, I Expect No Major Chaos

I think 2016’s shadow, even with things being as normal as ever (in the electoral sense at least!) in 2017, 2018, 2019, and earlier this year, has really clouded election analysts’ confidence. There’s some pressure (a lot of it unfair) that they “predict” these elections without error, and even though a lot of them know that there’s a lot of evidence for one result, they can’t help but to cover their bases and keep thinking “yeah but…”. I’m not trying to portray myself as any better; I myself, a data driven guy when it comes to election expectations, was asked by my card-carrying Democratic brother the other day how confident he should be Biden wins this. The polling data, the fundamentals, the historic trends, etc. say he’s at least a ninety percent likely winner, but my answer based on 2016 lingering in the back of mind was “seventy five percent.” And yet every Sunday I update my polling averages and find Biden ahead by bigger than 2016-miss margins.

So maybe this over three-thousand-word article comes back and bites me, maybe I’m a fool who hasn’t learned “nothing matters anymore”, but this is something that has been gnawing at me for weeks. Election surprises are rarely followed by another surprise in the following race, major polling errors are rarely followed by another major polling error, and election night (or week) chaos predictions usually end up being more built up than what happens in reality. There’s a reason the expected is…well, the expected.

I’m not asking you to not be vigilant about the race tightening. I’m not asking you to not keep in mind margins of error. I’m not (and Lord knows I can’t judge) asking you to not be worried about potential chaos come election night (or week). All I’m saying is, based on history, there is reason to be optimistic that what we expect will happen does happen, or that things end up running much smoother than the media might be hyping. After all, such surprises seem to come when we least expect them.